{"id":"W2919910741","doi":"10.1016/j.jss.2019.110486","title":"A machine-learning based ensemble method for anti-patterns detection","year":2019,"lang":"en","type":"preprint","venue":"Journal of Systems and Software","topic":"Software Engineering Research","field":"Computer Science","cited_by":5,"is_retracted":false,"has_abstract":false,"ca_institutions":"Concordia University; Polytechnique Montréal","funders":"Natural Sciences and Engineering Research Council of Canada; Canada Research Chairs","keywords":"Computer science; Ensemble learning; Machine learning; False positive paradox; Maintainability; Artificial intelligence; Aggregate (composite); Feature (linguistics); Source code; Program comprehension; Java; Class (philosophy); Data mining; Code (set theory); Pattern recognition (psychology); Software; Set (abstract data type); Software system","routes":{"ca_aff":true,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.002161665,0.000246946,0.0006597216,0.000470056,0.0001039119,0.0005003846,0.0006802636,0.0002793176,0.000001431721],"category_scores_gemma":[0.001222213,0.0002105896,0.0002501724,0.0001308683,0.000008518138,0.0001992992,0.0003497181,0.001007526,0.000001748625],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001335659,"about_ca_system_score_gemma":0.0002341193,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000123051,"about_ca_topic_score_gemma":0.00000358545,"domain_scores_codex":[0.9979888,0.000220925,0.0005791243,0.0003637235,0.0005498875,0.0002975519],"domain_scores_gemma":[0.9965175,0.001759257,0.0006462155,0.0004221933,0.0004987668,0.0001560295],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00007066118,0.00008475212,0.09483229,0.009594213,0.0004386046,0.0001391058,0.0009382358,0.8181559,0.001371022,0.0001030693,0.0003295978,0.07394259],"study_design_scores_gemma":[0.0009416512,0.0004578971,0.007741034,0.001862881,0.00004643315,0.000415819,0.00003328508,0.982785,0.0006690285,0.0001316532,0.004529173,0.0003860792],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01950764,0.002685138,0.9745956,0.00006320983,0.002585069,0.0004321797,0.00001787241,0.0001111455,0.00000211662],"genre_scores_gemma":[0.8486685,0.00008082989,0.150691,0.00001680159,0.0003640227,0.00002599405,0.00000345906,0.00004016562,0.0001092297],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.8291608,"threshold_uncertainty_score":0.8587594,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02462556380438579,"score_gpt":0.2921812808401903,"score_spread":0.2675557170358045,"validation_status":"score_only:v0-immature-baseline","note":"Baseline scores from an immature model (maturity gate not passed). Scores rank; they never assert a category."}}